Early detection of prostate cancer in Germany: a study using digital rectal examination and 4.0 ng/ml prostate-specific antigen as cutoff.

OBJECTIVE

While international screening studies for prostate cancer are by now almost reaching the estimated number of recruitments mandatory for the necessary power to investigate an effect on mortality of prostate cancer, no statistical figures on the detection of prostate cancer in Germany - apart from historical data before the use of prostate-specific antigen (PSA) are available. In order to generate a database and to investigate the diagnostic efficacy of the primarily practice-based urological care system, a case finding study designed as a nationwide longitudinal early detection trial was initiated.

METHODS

In one week in November 1997, 963 urologists prospectively examined 11,644 men between 45 and 75 years of age by digital rectal examination (DRE) and PSA with 4.0 ng/ml as cutoff. Data of family history and physical examination were collected by questionnaire. At this time participants were not aware of their PSA value. PSA was determined in the study center. Indication for sextant biopsy was a PSA value above 4.0 ng/ml or a suspicious lesion on DRE. Any indicated biopsy not performed had to be clarified. In a second questionnaire results of prostate biopsy, treatment and tumor status were documented.

RESULTS

The mean age of the study population was 62 years (median 62). The PSA median was 1.4 ng/ml with 82.8% presenting with < 4.0 ng/ml, 12.8% with 4-10 ng/ml and 4.4% with >10 ng/ml. From 1,115 men (47.7%) biopsied, 262 cancers were detected resulting in a detection rate of 23.5%. While 399 men refused biopsy, further investigation was not recommended in 387 men by their urologist, because prostatitis or benign hyperplasia was thought to be the cause for elevated PSA. From the 143 patients undergoing radical prostatectomy, 93 (65%) cancers were organ confined. T(1c) cancers with elevated PSA > 10 ng/ml could be treated with curative intent in 44% only. The positive predictive value (PPV) was estimated to be 16% for DRE alone (14/90), 17% for PSA alone (143/819) and 51% for the combination of both (105/206). In that cohort, use of age-adjusted PSA values and PSA density increased the PPV of PSA testing nonsignificantly.

CONCLUSIONS

Significant higher PPV indicated that utilizing a combination of both DRE and PSA is most effective in the early detection of prostate cancer. Unnecessary biopsies can be avoided using either age-adjusted PSA value or PSA density, but the PPV is not significantly changed and potentially curable cancer is missed in up to 25%. Given the substantial variability of the diagnostic approach despite the study design, uniform guidelines are necessary to ensure countrywide sufficient screening.